| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Technical Advance |
From the Laboratory of Molecular Biology, Institute of ClinicalPathology, Department of Pathology, University Hospital,Zürich, Switzerland
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Hence, the demonstration of the t(14;18) translocations in biopsy
samples has been widely used as an accessory diagnostic tool for the
identification of follicular lymphomas. Because of its relative
simplicity, polymerase chain reaction (PCR) is nowadays the most
commonly applied technique to detect this translocation. Other more
laborious approaches include Southern blotting, conventional
karyotyping, fluorescence in situ hybridization, and fiber
fluorescence in situ hybridization. In the past, the
standard PCR detection methods have been mainly based on observations
that the breakpoints on chromosome 18 cluster at two sites: in a
150-bp-long region located in the untranslated 3'-portion of the third
exon of the BCL2 gene2
and in a 500-bp DNA
stretch 
30 kb downstream of the BCL2 locus.8
These clusters have been named major breakpoint region (mbr) and minor
cluster region (mcr), respectively. On chromosome 14, breakpoints are
predominantly found within the joining elements (JH) of the
immunoglobulin heavy chain gene suggesting an aberrant recombination
process as the primary cause for the translocation. Because of these
apparently restricted breakpoint localizations, primers immediately 5'
to the mbr and the mcr clusters on chromosome 18 and in conserved
portions of the JH elements on chromosome 14 have been widely used for
standard PCR assays. The combination of these mbr or mcr primers with a
JH consensus primer gives rise to PCR products in the range of 150 and
500 bp, respectively, allowing a successful amplification even starting
from formalin-fixed and paraffin-embedded tissues.
In recent years, a large number of studies have analyzed t(14;18) translocations in follicular lymphoma.9 Surprisingly, the detection ratios of BCL2-IGH rearrangements varied greatly among these reports, likely being explained by the different methodologies used. From recent studies,10,11 using either a long-distance or an inverse PCR approach, it has become evident that the breakpoints on chromosome 18 are not only occurring within the mbr and mcr loci, but are often scattered between these two cluster regions. In consequence, these breakpoints escape the detection by conventional PCR techniques leading to a large number of false-negative results.
Although the long-distance PCR (LD-PCR) protocol of Akasaka and co-workers10 significantly enhances the detection rate of t(14;18) translocations in follicular lymphomas, it is not easily included in routine laboratory assays. Depending on the breakpoint localization, fragments up to 23 kb have to be amplified. In this size range the sensitivity of the assay is greatly affected by the quality of the DNA and minimal tissue infiltrates of lymphoma cells may be missed. Furthermore, exact breakpoint localization is often impossible because of the lack of sequence data covering the stretch between mbr and mcr.
We therefore isolated and sequenced the hitherto unknown genomic region between these two cluster regions, designed sets of new primer pairs for shorter t(14;18) target amplifications, and compared our improved PCR assay with the widely used standard PCR approaches.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
For this study, we have analyzed frozen tissues from 59 patients with follicular lymphoma admitted to University Hospital, Zürich, Switzerland. The tissues were subjected to standard histochemical and immunohistochemical analysis, including Bcl-2 immunostaining. Histological classification of follicular non-Hodgkin’s lymphoma was performed according to the World Health Organization classification.
DNA Extraction
DNA was isolated from snap-frozen tumor specimens using standard
protocols. Briefly, four to six 20-µm-thick cryostat sections were
digested overnight at 56°C in extraction buffer (100 mmol/L NaCl, 10
mmol/L Tris buffer, pH 8, 25 mmol/L ethylenediaminetetraacetic acid, pH
8, and 0.5% sodium dodecyl sulfate) containing 100 µg/ml proteinase
K (Roche Biochemicals, Rotkreuz, Switzerland). Remaining protein
debris was removed by phenol-chloroform extraction. Phase separation
was done in Phase Lock light tubes (Eppendorf, Basel, Switzerland)
according to the manufacturer’s instruction. Precipitation of the
purified DNA was performed with half a volume of 7.5 mol/L
NH4OAc and 3 volumes of ice cold ethanol. The DNA
was dissolved in TE buffer (10 mmol/L Tris buffer, pH 8, and 1 mmol/L
ethylenediaminetetraacetic acid, pH 8) and stored at 4°C until use.
The DNA concentration was measured by UV absorbance at 260 nm. The
integrity of the extracted DNA was tested by performing control LD-PCRs
with two different primer sets (A2-up/A2-low and B4-up/B4-low, Table 1
).
|
Two genomic fragments covering the entire stretch between mbr and
mcr were amplified using two LD-PCRs (primer pairs A2-up/A2-low and
B4-up/B4-low, Table 1
). These primer pairs were designed on the basis
of previously published sequences (M14745, AB010948, AB010949, sequence
from Ngan and colleagues12
). The cycling conditions
included an initial denaturation for 1 minute at 94°C followed by 14
cycles with 98°C denaturing for 20 seconds and 20 minutes
annealing/primer extension at 68°C plus 16 cycles consisting of 20
seconds of denaturing at 98°C and annealing/primer extension at
68°C for 20 minutes with 15-second increments per cycle; final
extension 10 minutes at 72°C. Each reaction mixture (50 µl)
contained 100 to 200 ng DNA in 1x PCR reaction buffer (TaKaRa) with
0.4 µmol/L primers, 0.4 mmol/L dNTPs, 2.5 mmol/L
MgCl2, and 2.5 U LA Taq-Polymerase
(TaKaRa, Otsu, Japan). The reactions were overlaid with one drop of
mineral oil. PCR amplification was performed in a Cetus/Perkin-Elmer
thermocycler 480.
The LD-PCR products were either directly ligated into the pGEM-T-easy vector (Promega, Madison, WI) (13.3-kb 5'-fragment) or subcloned into the pBluescript SK+ vector (Stratagene, La Jolla, CA) after digestion with either EcoRI, SacI, or PstI restriction enzymes (11.5-kb 3'-fragment). For all subcloning experiments the recombination-deficient Escherichia coli strain Sure (Stratagene) was used as host. Direct sequencing of the PCR products and/or the subcloned fragments was done on an ABI 377 sequencer using Big Dye terminator kits (Applied Biosystems, Rotkreuz, Switzerland) and custom primers.
t(14;18)-Specific Polymerase Chain Reactions for Standard and Long Target Amplification
Conventional PCRs were essentially done according to the protocol of Liu and co-workers.21 To obtain optimal primer annealing (primer analysis with Oligo 5.0 software), the mbr and mcr standard primers were shortened at their 3' end by two nucleotides. PCR conditions for standard PCR were as follows: initial Taq activation/DNA denaturation for 3 minutes at 96°C, followed by 35 cycles consisting of 2 minutes of denaturation at 94°C, 1 minute annealing at 58°C, and 2 minutes of primer extension at 72°C. The 50-µl reactions contained 100 ng DNA, 1x reaction buffer II (Perkin Elmer, Rotkreuz, Switzerland), 0.2 µmol/L each primer, 0.2 mmol/L dNTPs, 2.5 mmol/L MgCl2, and 2.5 U Amplitaq Gold (Perkin Elmer). PCR amplification was performed in a Perkin-Elmer thermocycler 9600.
For LD-PCR new primer pairs were designed with the Oligo 5.0 software
(MedProbe, Oslo, Norway) on the basis of our novel sequence data (Table 1)
. These primer sets were supplemented with previously published
LD-PCR primers.10
Identical LD-PCR-cycling conditions as
described before were applied. Each DNA sample was analyzed at least
two times in independent experiments.
To avoid contamination, all PCRs were prepared in a laminar flow hood and pipette tips with aerosol filters were used. Each run included a negative control, in which the DNA solution was replaced by water, and a positive control amplification of a patient’s sample with confirmed t(14;18) translocation in the corresponding region. Furthermore, the DNA quality of each extract was tested in an internal control amplification using primers specific for the versican13 (vExon3up: CAACGATGCCTACTTTGCCACCC/vExon3low: ATCCCGTACATGACGTCACAGCG; standard PCR) or the BCL2-gene (A2-up/A2-low and B4-up/B4-low; LD-PCR).
t(14;18) Breakpoint Identification
Ten-µl aliquots of the PCR reactions were analyzed by agarose gel electrophoresis (1% agarose gels for standard PCR products and 0.5% for LD-PCR products) followed by ethidium bromide staining. PCR products were excised from the gels and isolated with an agarose gel extraction kit (Qiagen, Basel, Switzerland).
The presence of a t(14;18) translocation was verified by sequencing each patient’s sample with the corresponding PCR primers and JH-specific sequencing primers for exact breakpoint identification. The amplification products were either sequenced directly or after subcloning into pGEM-T vector (Promega). Sure bacteria (Stratagene) were used as host strain. The resulting sequences were aligned with the IGH gene locus (accession number X97051), with the BCL2 (M14745), mcr,12 and the novel DNA sequence between these two clusters (AF325194, AF325195) using Blast2 (www.ncbi.nlm.nih.gov/BLAST/).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
To improve PCR-based t(14;18) translocation detection
assays and to allow an unambiguous identification of specific
breakpoints between mbr and mcr, we have isolated and sequenced an
25-kb-long hitherto uncharacterized DNA stretch downstream of the
BCL2 gene locus. For this purpose, we performed two LD-PCRs
with primer pairs (A2-up/A2-low and B4-up/B4-low; Figure 1A
) designed on the basis of the
previously published sequences in this region (GenBank accession
numbers M14745, AB010948, AB010949, and from Ngan and
colleagues12
). In this way we obtained a 13.3 kb-
(A2-up/A2-low) and an 11.5-kb (B4-up/B4-low) amplification product,
respectively, bridging the entire unknown sequence between mbr and mcr
(Figure 1B)
.
|
1-kb-long CA microsatellite-containing element 22 kb downstream of
the mbr locus, was inaccessible to exact sequence analysis. Hence, the
novel genomic sequence downstream of the BCL2 was submitted
to GenBank in two parts: 1) the sequence stretch between mbr and the CA
dinucleotide repeat (accession number AF325194), and 2) the stretch
between the CA repeat and the mcr locus (accession number AF 325195).
No evidence for the presence of a gene in this DNA stretch immediately
downstream of the BCL2 locus could be found. Detection of t(14;18) Translocation Using Standard PCR Analysis
In a study of 59 patients with histologically
confirmed follicular lymphoma we initially screened for the presence of
a t(14;18) translocation applying the widely used standard PCRs with
mbr/JH and mcr/JH primer pairs (Table 1)
. In this way we could
demonstrate a translocation in 36% of the samples (21 patients)
including 19 samples (32.2%) with breakpoints within the mbr cluster
and 2 samples (3.4%) with involvement of the mcr locus. The size of
the PCR products containing breakpoints in the mbr ranged from 180 to
275 bp. The amplicons of the two mcr samples were 580-bp and 630-bp
long. Representative results from the standard t(14;18) PCRs are shown
in Figure 2
. In addition an 500-bp
amplification product was obtained from patient 40 in both the mbr and
the mcr reaction. Further examination of this sample revealed that the
product resulted from an amplification with JH consensus primer alone
(for exact breakpoint identification see separate paragraph below).
|
Because the standard PCRs detect only breakpoints within the two
cluster regions, we have performed a second screening round using a
novel LD-PCR approach. Based on our new sequence data from the genomic
locus 3' of the BCL2 gene, we established two sets of LD-PCR
assays (Figure 3
, set A and B). With
these assays, we covered the entire region between the third exon of
the BCL2 gene and mcr allowing the detection of breakpoints
outside of mbr and mcr. The new primers were positioned in regular
intervals along the 31-kb-long sequence stretch to limit the size of
the PCR products to less than 15 kb (set A) or 9 kb (set A and B),
respectively.
|
), allowing an approximate positioning
of the translocation breakpoint between mbr and mcr.
|
Exact Characterization of Translocation Breakpoints by Sequence Determination
Exact breakpoint localization through sequence determination and
comparison confirmed in 19 samples a translocation involving the mbr
region (mbr as defined by Bakhshi and colleagues14
) and in
2 samples a breakpoint within the mcr (determined by Ngan and
colleagues12
) (Figure 5
,
Table 2
). Twenty patients with a
follicular lymphoma displayed breakpoints scattered throughout the
30-kb region spanning mbr and mcr and in one case (patient 42) a
breakpoint 800-bp 5' from the mbr, but 1.5 kb downstream from the
bcl-2 translation stop codon was detected.
|
|
5.3
kb and 6.2 kb downstream of the mbr locus (three samples each). In
addition, a very high prevalence for chromosome 18 breakpoints was
observed 19-kb 3' from the mbr. Seventeen percent of the (t14;18)
translocations (seven samples) could be detected in an 200-bp-long
sequence portion, we now call the "intermediate cluster region"
(icr). In comparison, we observed in our study an involvement of mbr or
mcr in 45% and 5% of the translocations, respectively. Interestingly,
the breakpoints in this new cluster were identical in patients 36, 37,
and 38 and differed only by 1 bp in patient 35. Despite the
similarities on the 3'-BCL2 side, N-nucleotide additions and
breakpoints in the IGH locus on chromosome 14 were unique in
each of these samples. A representative LD-PCR analysis of a patient
with a breakpoint in this icr is shown in Figure 4
(sample 36). On chromosome 14 the translocation involved in six samples the IGH joining element JH4 (14%), five times JH5 (12%), and in 23 patients the JH6 segment (55%). In an additional two samples (4.75%) the translocation occurred further downstream lacking all JH segments on chromosome 14+, but leaving the Eµ enhancer intact. Twice we observed a link between BCL2 or 3' sequences thereof to partially rearranged (D-J) diversity elements (4.75%) and another two times the 3'-recombination signal sequence (3'-RSS) of D3-9 participated in the joint at the breakpoint (4.75%). Finally, more complex rearrangements including a t(14;18) translocation and a partial inversion of the JH region were found in two instances (4.75%). At these breakpoints some of the joining elements were aberrantly arranged in opposing directions. In consequence a conventional PCR with JH consensus primers alone yielded in one analysis a short amplification product (patient 40).
All except for one breakpoint fusion sequences included on chromosome 14+ variable numbers of N-nucleotide additions. No such modifications were present in a rather unusual translocation joining a sequence 5' to the mbr directly with a DNA stretch 3' to the JH segments (patient 42).
Improved Detection of t(14;18) Translocations in Formalin-Fixed Tissues
Our follicular lymphoma study demonstrates that breakpoints occur
more frequently within the icr relative to translocations involving the
mcr. Nevertheless, neither the widely used protocols for standard PCRs
nor our LD-PCRs are suitable to detect breakpoints in the icr in DNA
extracts from tissues that have been formalin-fixed and
paraffin-embedded. To make this novel cluster accessible to such
analysis, we have designed a new primer (s-icr) to supplement the
widely used standard PCRs with s-MBR and s-mcr primers (Table 1)
. The
combination of s-icr with the JH consensus primer detected in six
instances breakpoints located in the icr also starting from
formalin-fixed tissue samples (for example see Figure 6
). Only in one case with a breakpoint
downstream from JH6 on chromosome 14 (patient 35), the PCR with the
s-icr/JH primer pair failed to detect a translocation because of the
absence of the JH primer-annealing site. Hence, by including the
s-icr/JH primer pair in the standard PCR assay we could significantly
increase the overall t(14;18) detection rate in DNA extracts from
formalin-fixed tissues from 36 to 46% (Figure 7)
.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Our study makes evident that only half of the breakpoints localize within the commonly analyzed clusters mbr and mcr. The other breakpoints are widely spread throughout the sequence stretch between mbr and mcr, nevertheless displaying some degree of clustering. This is most pronounced in the icr, where we have observed more than three times more breakpoints than in the mcr. These data are supported by the findings of Willis and colleagues11 and Akasaka and colleagues,10 who also identified a few breakpoints in this region.
The observation that a significant proportion of the chromosome 18
breakpoints localize far distant from mbr and mcr explains why standard
PCR assays detect fewer t(14;18) translocations than other techniques.
Although the sensitivity of these assays can be improved by including
primers for the icr, approximately a third of the t(14;18)
translocations remain inaccessible to these conventional PCR methods
(Figure 7)
. Nevertheless, in situations in which only formalin-fixed
and paraffin-embedded biopsies are available, standard PCRs with
s-MBR/JH, s-mcr/JH and s-icr/JH primer combinations may be the only
detection technique available, because of extensive DNA degradation in
fixed samples.
The 71% proportion of t(14;18) translocations we detect with our LD-PCR protocol in follicular lymphomas correlates well with results from cytogenetic and Southern blot studies from the United States and is significantly higher than similar data from Europe and the Far East.9 Because we found in each of the LD-PCR-positive samples distinct translocation sequences, false-positive results because of unspecific amplification and contamination can be excluded. We assume that the difference in sensitivity between European and North American Southern blot studies may rather arise from methodological differences than from ethnic and/or geographic peculiarities. Theoretically, all breakpoints in our study could have been detected in Southern blots using an mbr and an mcr probe together with a BamHI restriction, despite the fact that BamHI does not cover the entire stretch between mbr and mcr. However, some of the breakpoints would have given rise to fragments larger than 30 kb being well beyond the separation capacity of conventional agarose gel electrophoresis. Even more significant, 8 and 16 breakpoints of our study would have been missed in protocols using HindIII and EcoRI digestions, respectively. Because not all of the European studies included BamHI in the Southern blot analysis, a major proportion of the breakpoints may have remained undiscovered.
Exact t(14;18) breakpoint characterization by directly sequencing the
PCR products not only allows the verification of the amplification
specificity, it may also give insight into the mechanisms that lead to
these aberrant gene rearrangement events. Various models have been
presented, all proposing an involvement of the VDJ recombination
machinery in the translocation. Whereas a participation of RAG-1 and
RAG-2 in the formation of double-strand breaks in the IGH
gene on chromosome 14 is undisputed, there are still doubts about the
mechanisms generating the breakpoints on chromosome 18. Cryptic
recombination signal sequences (RSS)17
within or near the
BCL2 gene locus or RAG-mediated transposition
events18
may be involved in the translocation process. In
our study, we have found only one breakpoint localization with a
potentially cryptic RSS 3.3 kb downstream of the BCL2
gene. It contains a putative nonamer and heptamer sequence with a 22-bp
spacer still conforming to the 12/23 rule.19
At this site,
the chromosome 18 sequence of three patients either ended directly
after the heptamer (patients 29 and 30) or 1-bp 3' thereof (patient
31). Recently, the same breakpoint location has been identified in a
study by Vandraager and colleagues.20
In patients
30 and 31 of our study, these putative RSS were linked to the 3'-RSS of
the D3-9 element in the IGH locus. Hence, these
translocations may have resulted from an aberrant signal joining,
despite the unusual presence of N-nucleotides in both joints. In one
instance (patient 22) we found also a breakpoint that partly conforms
to the transposition model, in which one signal end attacks the
double-stranded DNA of another chromosome. The addition of
N-nucleotides as present in our sample is, however, rather
uncharacteristic for such a mechanism. The high variability of
breakpoint localization and junctional sequences observed in the rest
of the samples argues against a single aberrant process causing all
t(14;18) translocations. It seems rather likely, that different defects
in the RAG-driven immunoglobulin rearrangement and eventually even
joining of randomly broken chromosomal ends may lead to similar
translocations becoming only manifest as a lymphoma after an extended
cellular selection.
In conclusion, our protocol has greatly improved the detection rate of t(14;18) translocations in follicular lymphomas using a relatively simple PCR technique. This method is suitable for routine diagnosis in a standard molecular pathology laboratory. Our LD-PCR and sequencing approach is rapid, highly specific, and sensitive and could eventually replace Southern blotting analysis commonly used for samples in which high-molecular weight DNA is available. By determining a large sequence stretch downstream from the BCL2 gene locus, we have furthermore provided the basis for the exact characterization of a series of novel breakpoints distant from mbr and mcr, finally allowing the development of sensitive patient-specific PCR assays for the monitoring of therapeutic success and the detection of minimal residual disease.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Present addresses of A. A.-H.: HNO-Universitätsklinik Freiburg, Killianstr. 5, 79106 Freiburg, Germany; and of B. H.: Medizinische Abteilung, Spital Limmattal, Urdorferstr. 100, 8952 Schlieren, Switzerland.
Accepted for publication December 20, 2001.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Related articles in Am J Pathol:
This article has been cited by other articles:
 |
|
 |
|
  C. M. McHale, Q. Lan, C. Corso, G. Li, L. Zhang, R. Vermeulen, J. D. Curry, M. Shen, R. Turakulov, R. Higuchi, et al. Chromosome Translocations in Workers Exposed to Benzene J Natl Cancer Inst Monographs, July 1, 2008; 2008(39): 74 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. K. Weinberg, W. Z. Ai, M. R. Mariappan, C. Shum, R. Levy, and D. A. Arber ''Minor'' BCL2 Breakpoints in Follicular Lymphoma: Frequency and Correlation with Grade and Disease Presentation in 236 Cases J. Mol. Diagn., September 1, 2007; 9(4): 530 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A Catherwood, D. Gonzalez, C. Patton, E. Dobbin, L. Venkatraman, and H D. Alexander Improved clonality assessment in germinal centre/post-germinal centre non-Hodgkin's lymphomas with high rates of somatic hypermutation J. Clin. Pathol., May 1, 2007; 60(5): 524 - 528. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L Venkatraman, M A Catherwood, A Patterson, T F Lioe, W G McCluggage, and N H Anderson Role of polymerase chain reaction and immunocytochemistry in the cytological assessment of lymphoid proliferations J. Clin. Pathol., November 1, 2006; 59(11): 1160 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P J Batstone and J R Goodlad Efficacy of screening the intermediate cluster region of the bcl2 gene in follicular lymphomas by PCR J. Clin. Pathol., January 1, 2005; 58(1): 81 - 82. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Raghavan, S. Houston, B. G. Hegde, R. Langen, I. S. Haworth, and M. R. Lieber Stability and Strand Asymmetry in the Non-B DNA Structure at the bcl-2 Major Breakpoint Region J. Biol. Chem., October 29, 2004; 279(44): 46213 - 46225. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Roulland, P. Lebailly, Y. Lecluse, M. Briand, D. Pottier, and P. Gauduchon Characterization of the t(14;18) BCL2-IGH Translocation in Farmers Occupationally Exposed to Pesticides Cancer Res., March 15, 2004; 64(6): 2264 - 2269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Barrans, P. A.S. Evans, S. J.M. O'Connor, R. G. Owen, G. J. Morgan, and A. S. Jack The Detection of t(14;18) in Archival Lymph Nodes: Development of a Fluorescence in Situ Hybridization (FISH)-Based Method and Evaluation by Comparison with Polymerase Chain Reaction J. Mol. Diagn., August 1, 2003; 5(3): 168 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sanchez-Beato, A. Sanchez-Aguilera, and M. A. Piris Cell cycle deregulation in B-cell lymphomas Blood, February 15, 2003; 101(4): 1220 - 1235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Biagi and J. F. Seymour Insights into the molecular pathogenesis of follicular lymphoma arising from analysis of geographic variation Blood, May 29, 2002; 99(12): 4265 - 4275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Aster and J. A. Longtine Detection of BCL2 Rearrangements in Follicular Lymphoma Am. J. Pathol., March 1, 2002; 160(3): 759 - 763. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
